TRACE: Capability-Targeted Agentic Training

Large Language Models (LLMs) deployed in agentic environments must exercise multiple capabilities across different task instances, where a capability is defined as performing the actions required for the successful completion of some subset of tasks. Many existing approaches either rely on synthetic training data that is not targeted to the model's actual capability deficits in the target environment or train directly on the target environment, where the model needs to implicitly learn the capabilities across tasks. We introduce TRACE (Turning Recurrent Agent failures into Capability-targeted training Environments), an end-to-end system for environment-specific agent self-improvement. TRACE contrasts successful and failed trajectories to automatically identify frequently lacking capabilities, synthesizes a targeted training environment for each capability that rewards its successful execution, and trains a LoRA adapter via RL on each synthetic environment, routing to the relevant adapter at inference. Empirically, TRACE generalizes across different environments, improving over the base agent by **+14.1 points** on $\tau^2$-Bench (customer service) and **+7 perfect scores** on ToolSandbox (tool use), outperforming the strongest baseline by **+7.4 points** and **+4 perfect scores**, respectively. Given the same number of rollouts, TRACE scales more efficiently than baselines, outperforming GRPO and GEPA by **+9.2** and **+7.4** points on $\tau^2$-Bench, respectively.